Roll’s conceptual discussion is supported by empirical research. Andy Naranjo and Alden Toevs (2002) conclude that portfolio purchases affect rates and volatility of conforming and nonconforming loans. Akash Deep and Dietrich Domanski (2002) conclude that expansion of the GSEs’ portfolios has absorbed, or enabled, a significant portion of new mortgages and refi- nancings. And Gloria Gonzalez-Rivera (2001), Stuart Gabriel (2001) and Douglas McManus and Buchi Ramagopal (2006) find that the GSEs’ retained portfolio activities increase liquidity in the secondary mortgage market.
It is perhaps the case that the benefits of the GSEs’ portfolios have not been quantified in terms of a single metric to compare with the risks of default. Yet, given that the risk of default is low and the benefits are numerous and significant, one should be wary of any proposal to curtail the GSEs’ retained portfolios.
Revisiting the Net Benefits of Freddie Mac and Fannie Mae
This is not to say the risks should be ignored. Certainly, continual improvement in the sophisti- cation of OFHEO’s risk-based-capital test is warranted, as is an insistence that the GSEs continually pass this test. Even greater transpar- ency in the GSEs’ portfolios and other activities would also reduce the risk and, perhaps as importantly, contribute to the public’s confidence in the GSEs’ operations.
The Policy Debate Over GSE Activities